Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

safe voltages

Options
  • 08-02-2004 12:45pm
    #1
    Registered Users Posts: 6,560 ✭✭✭


    Hi want to overclock my 2500+ a little more currently sitting at 200*11 @ 1.75vcore. cooling as outlined in sig, i've upped the vcore to 1.85 and started running prime, temps are currently ok 45c or so with window open, just wondering if this type of vcore is "safe" for me chip?


Comments

  • Registered Users Posts: 771 ✭✭✭Sir Random


    Why did you raise the vcore? was it not stable at 200x11, 1.75v?

    You should only really need 1.85v if you're trying for speeds like 2.4+Ghz (~220x11), or is that what you're doing?


  • Registered Users Posts: 6,560 ✭✭✭Woden


    aye its well stable at 200*11 at 1.75 was going for 200*11.5 which turns out isn't prime 95 stable


  • Registered Users Posts: 771 ✭✭✭Sir Random


    Can't you lock the AGP/PCI with that mobo and try upping the fsb?


  • Registered Users Posts: 6,560 ✭✭✭Woden


    sure the i presume the agp/pci is locked already? i've never touched it and i've never had any problems even when i've upped the fsb over 200, i can get it to boot into windows up til about 218 mhz fsb,

    i'm not sure what voltage gives fsb stability, is it the core voltage i thought that helped more with mulitplier increase, what about the chipset voltage thru the northbridge, mines at 1.6V can be increased to 1.7v


  • Registered Users Posts: 771 ✭✭✭Sir Random


    vcore gives fsb & multiplier stability (cpu stability)
    More NB voltage helps stability when transferring data at high bandwiths (cpu & ram both o/c'd) but I doubt you'd need more than 1.6v unless you're using a high fsb with 1:1 ratio

    Did you get 218x11 at 1.75v? If so, you should get 220x11 at 1.85v. Then again, you're initial question was about the safety of running a higher vcore which is, unfortunately, impossible to predict.


  • Advertisement
  • Registered Users Posts: 2,754 ✭✭✭Col_Loki


    There is a huge ammount of info on it here - http://www.ocforums.com/vb/vb/showthread.php?s=&postid=2168559 , might be helpfull.

    Your temps definetly arent a problem, your cooling is top notch. It looks like its the chip holding you back on the overclock :( . Am sure you know this already but.... the abit undervolts from the setting in the bios (ie when you set to 1.75v its really around 1.7v to the chip) , also the temps on the abit NF7 are roughly 8 degrees higher than the ones you will get on a simular setup with say an Asus board....... so your temps are very good.

    The newer Xp2500+'s dont seem to have much of a problem with high Vcore, some of the older ones wouldnt post if you went over 1.85-1.9v ......it would be perfectly stable at say 2400mhz @ 1.85v, set it to 1.875 and 2400mhz (or a little more) and its not prime stable. The one im using now doesent have a problem with 2.05v and up .

    Safe voltage is very hard to to estimate, the link above gave me a clearer understanding of it and i poped my Vcore up to 2.05v :) . Im currently running 1.9v @ 2400mhz , watercooling set to lower speeds (quietness)...... i have no problem with this as my temps are roughly 50-52 under load, and 44 idle. (i worry more about temps than voltage, dont know why).


  • Registered Users Posts: 6,560 ✭✭✭Woden


    thanks for the replies folks,

    just messing around with it, i have the wierdest ram i tell ya, computer boots into windows at 220*9 @ 1.85v and 2-3-3-7 prime 95 fails in about 2 seconds :), so i lowered it to 216 and loosened the ram timings to 3-4-4-11 and it won't even post lol, got back into the bios and tightened the timings to 2-3-3-7 and i'm back in windows running prime which is more stable then it was previously.

    gonna mess around with it and get some more and get back to ya's god i should be studying :)


  • Registered Users Posts: 15,815 ✭✭✭✭po0k


    Dunno if it's been mentioned already, but raising voltage should only be used as a last resort.
    Getting better cooling should be your first aim as raising the VCore will just produce more heat anyways and has a negative effect on the life of the semiconductor gates in the chip.

    1.85 is quite high aswell.
    a ~10% increase is really the most you should go for safety.
    Course, if depends on the particular chip aswell and can be rather hit or miss.


  • Registered Users Posts: 6,560 ✭✭✭Woden


    well i have fairly good air cooling even with upping the vcore the the temperatures have remained reasonable, i've justed slipped back to 200*11 @ 1.75vcore for the moment as i'm not seeing massive performance increases in performance for the small overclock with large a large increase in vcore


  • Hosted Moderators Posts: 2,559 ✭✭✭Tazzle


    On that note, anyone got any advice as to why my athlon 1900xp on an abit kx7-333 will not get past boot at any value higher than the default 133, it'll just lock up at some point through the loading of windows, and anything above 144 and it will not even post.


  • Advertisement
  • Registered Users Posts: 6,560 ✭✭✭Woden


    does your motherboard lock the agp/pci thingy? what happens if you up the multi?


  • Hosted Moderators Posts: 2,559 ✭✭✭Tazzle


    The multiplier on my palomino is locked, and I can't be fúcked fixing that. As for my motherboard lock the agp/pci thingy? What are you on about? the divider for the speeds? if so it's set to 4:2:1, 3:2:1 makes it lock up after about 2 hours on normal settings, 5:2:1 has no effect on stability.


  • Registered Users Posts: 6,560 ✭✭✭Woden


    lol i'm not sure to tbh thats why i put the thingy on it, afaik its something to do with the speed the agp and stuff runs at some boards don't lock the frequency or something which means that as you up the fsb they get messed with, resulting in problems with graphics cards data corruption from you hdd and stuff, certain motherboard lock it other don't i think.

    someone else will be able to give you a better explaination

    edit heres something small about it
    http://forums.extremeoverclocking.com/showthread.php?forumid=100&threadid=81790


  • Hosted Moderators Posts: 2,559 ✭✭✭Tazzle


    Example:
    4:2:1, this means the agp runs at half the fsb and the pci runs at 1/4 the fsb
    5:2:1, this means the agp runs at 2/5ths the fsb and the pci runs at 1/5th the fsb


  • Registered Users Posts: 6,560 ✭✭✭Woden


    ah ok we perhaps you don't have a lock then and you want the agp at 66mhz at the pci at 33mhz i think


  • Hosted Moderators Posts: 2,559 ✭✭✭Tazzle


    If it was a problem with the agp or pci then at 5:2:1 I would be able to run at an fsb of 165 without stressing them, HOWEVER, as previously stated, I cannot. Not being able to increase my fsb by even 5mhz


  • Registered Users Posts: 6,560 ✭✭✭Woden


    no not necessarily as that is quite a big overclock from 133-165 and your chip, board, ram whatever may not be able to handle it,


  • Hosted Moderators Posts: 2,559 ✭✭✭Tazzle


    Have you caught stupid?


  • Registered Users Posts: 6,560 ✭✭✭Woden


    look mate i'm not looking for your smark ass remarks i was trying to help out do your own ****ing research into your own problems and start your own ****ing thread.

    you have could have a crap chip, a crap psu, crap cooling, high load temp anything that could stop you overclocking plus a motherboard that doesn't allow you to lock the agp/pci frequencies


  • Hosted Moderators Posts: 2,559 ✭✭✭Tazzle


    Woah, you should definitely have that checked out, it might be contagious.


  • Advertisement
  • Registered Users Posts: 6,892 ✭✭✭bizmark


    tazzle stop being a troll ffs


  • Registered Users Posts: 6,560 ✭✭✭Woden


    yeah whatever tazzle

    back on track anyway, dropped it back to 200*11 @ 1.75 and keep tightening up the ram timings will go for 2-2-2-5 seems to get through a pifast benchmark fine for me which is one of the things that usually crashes for me if the memory is an issue.

    with the new cats 3dmax gets about 17200, what do yous think about overclocking the gfx card?


Advertisement